CN113743386A

CN113743386A - Lane line generation method, apparatus, device, and storage medium

Info

Publication number: CN113743386A
Application number: CN202111306338.6A
Authority: CN
Inventors: 肖童星
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2021-11-05
Filing date: 2021-11-05
Publication date: 2021-12-03
Anticipated expiration: 2041-11-05
Also published as: CN113743386B

Abstract

The embodiment of the application discloses a lane line generation method, a device, equipment and a storage medium, wherein the method relates to the technical field of automatic driving and artificial intelligence, and the method comprises the following steps: acquiring an actual side lane line of a first road section and a center lane line of a second road section; determining the width information of the second road section according to the actual side lane line of the first road section; extracting a center position point for reflecting the contour of the center lane line of the second road section from the center lane line of the second road section; determining a side position point for reflecting the outline of the virtual side lane line of the second road section according to the center position point and the width information of the second road section; and generating a virtual side lane line of the second road section according to the side position point. The accuracy and the efficiency of generating the side lane line can be improved through the method and the device.

Description

Lane line generation method, apparatus, device, and storage medium

Technical Field

The present application relates to the field of automatic driving technologies and artificial intelligence technologies, and in particular, to a lane line generation method, apparatus, device, and storage medium.

Background

The autopilot technology involves knowing the surrounding traffic conditions through video cameras, radar sensors and laser range finders and navigating the road ahead through a detailed map (a map collected by a person driving a car). In practice, it is found that if a side lane line (such as a left lane line and a right lane line) is absent in a front road, a navigation error is easily caused, and further, a user cannot efficiently reach a destination address. It can be seen that in the automatic driving scene, the lane line is taken as important static semantic information, and has great significance for vehicle navigation.

At present, side lane lines are mainly added to roads lacking side lane lines in a manual mode, and due to the influence of subjective factors, the accuracy and the efficiency of the method for generating the side lane lines are low.

Disclosure of Invention

The technical problem to be solved by the embodiments of the present application is to provide a lane line generation method, apparatus, device and storage medium, which can improve the accuracy and efficiency of generating a side lane line.

An embodiment of the present application provides a lane line generation method, including:

acquiring an actual side lane line of a first road section and a center lane line of a second road section; the second road section and the first road section have an adjacent relation, and the second road section does not contain an actual side lane line;

determining the width information of the second road section according to the actual side lane line of the first road section;

extracting a center position point for reflecting the contour of the center lane line of the second road section from the center lane line of the second road section;

determining a side position point for reflecting the outline of the virtual side lane line of the second road section according to the center position point and the width information of the second road section;

and generating a virtual side lane line of the second road section according to the side position point.

An aspect of an embodiment of the present application provides a lane line generating device, including:

the acquisition module is used for acquiring the actual side lane line of the first road section and the center lane line of the second road section; the second road section and the first road section have an adjacent relation, and the second road section does not contain an actual side lane line;

the determining module is used for determining the width information of the second road section according to the actual side lane line of the first road section;

an extraction module, configured to extract a center position point that reflects a contour of a center lane line of the second road segment from the center lane line of the second road segment;

the determining module is further configured to determine, according to the center position point and the width information of the second road segment, a side position point for reflecting an outline of a virtual side lane line of the second road segment;

and the generating module is used for generating the virtual side lane line of the second road section according to the side position point.

One aspect of the present application provides a computer device, comprising: a processor and a memory;

wherein, the memory is used for storing computer programs, and the processor is used for calling the computer programs to execute the following steps:

An aspect of the embodiments of the present application provides a computer-readable storage medium, where a computer program is stored, where the computer program includes program instructions, and the program instructions, when executed by a processor, perform the following steps:

An aspect of an embodiment of the present application provides a computer program product, which includes a computer program/instruction, and when executed by a processor, the computer program/instruction implements the steps of the method described above:

In the application, the computer device may obtain an actual side lane line of the first road segment and a center lane line of the second road segment, where the first road segment and the second road segment have an adjacent relationship, and the second road segment does not include the actual side lane line, that is, the second road segment does not lack the side lane line; then, width information of the second road section is determined by the actual side lane line of the first road section, and a center position point reflecting the contour of the center line of the second road section is extracted from the center lane line of the second road section. Because the side lane line of the second road section has similarity with the center lane line of the second road section, that is, the profile of the side lane line of the second road section has similarity with the profile of the center lane line of the second road section; therefore, a side position point for reflecting the virtual side lane line of the second road section may be determined by the center position point and the width of the second road section, and the virtual side lane line of the second road section may be generated based on the side position point. Therefore, in the application, the virtual side lane line of the second road section can be automatically generated according to the central position point on the central lane line of the second road section and the width information of the second road section, the process of generating the virtual side lane line is not needed to be participated manually, and the accuracy and the efficiency of generating the side lane line are improved. The scheme can be applied to scenes such as automatic driving, auxiliary driving and the like, and accurate navigation information can be provided for the vehicle according to the virtual side lane line.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic diagram of an architecture of a lane line generation system provided in the present application;

fig. 2 is a scene intention of a server in a lane line generation system provided by the present application to generate a virtual side lane line;

fig. 3 is a schematic flow chart of a first lane line generation method provided in the present application;

FIG. 4 is a schematic illustration of a scenario provided herein in which a link width of a second link is determined based on an actual side lane line of a first link;

fig. 5 is a flowchart of a second lane line generation method provided in the present application;

FIG. 6 is a schematic view of a scenario provided herein for obtaining side position points for reflecting a contour of a virtual side lane line of a second road segment;

fig. 7 is a schematic flowchart of a first process for smoothing candidate side lane lines according to the present application;

fig. 8 is a schematic view of a second scenario for smoothing candidate side lane lines according to the present application;

fig. 9 is a schematic view of a scene where a virtual side lane line is generated after encryption processing is performed on a center position point according to the present application;

fig. 10 is a schematic structural diagram of a lane line generation apparatus according to an embodiment of the present application;

fig. 11 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The method is mainly used for the virtual road section, and the virtual road section can be a road section which only has a central lane line and lacks a side lane line. In one embodiment, a virtual road segment may refer to a road segment that is actually (i.e., offline) marked with only a center lane line, and the road segment that is not actually marked with a side lane line, and a map that includes the road segment (i.e., online) is also marked with no side lane line of the road segment. In another embodiment, the virtual road segment may refer to a road segment actually marked with a central lane line and a side lane line, but a map containing the road segment is not marked with the side lane line of the road segment. The lack of a side lane line may refer to any of the following three cases: a. lack of left lane line, b, lack of right lane line, c, lack of left lane line and right lane line.

The map herein may refer to a map of a lane line including a road segment, e.g., the map may refer to a high-precision map serving an autonomous driving system. The high-precision map is also called an automatic driving map and a high-resolution map, and is a new map data normal form for an automatic driving automobile. The absolute position accuracy of the high-precision map is close to 1m, and the relative position accuracy is in the centimeter level and can reach 10-20 cm. The method accurately and comprehensively represents road characteristics (such as lane line characteristics), requires higher real-time performance, and is the most remarkable characteristic of a high-precision map. In addition, the high-precision map records specific details of driving behaviors, including typical driving behaviors, optimal acceleration points and braking points, road condition complexity, labels on signal receiving conditions of different road sections and the like.

In an autopilot application scenario, it is common to provide navigation for a vehicle based on the lane lines of a road segment, and thus, if the road segment lacks a side lane line, navigation errors are easily caused. Based on this, the present application provides a lane line generation method, in which, first, an actual side lane line of a first road section and a center lane line of a second road section are obtained; the first road section is a road section with an actual side lane line, the second road section is a road section without the actual side lane line, and the actual side lane line is an actually existing side line in the second road section under the line. Then, the width information of the second road section is determined according to the actual side lane line of the first road section, a center position point for reflecting the outline of the center lane line is extracted from the center lane line of the second road section by utilizing the machine learning technology in artificial intelligence, and the side position point for reflecting the outline of the virtual side lane line of the second road section is determined according to the center position point and the width information of the second road section. Generating a virtual side lane line of the second road section through the side position points; a virtual side lane line may refer to a side lane line that is missing in the second road segment that is off-line and that is added to the second road segment that is on-line (in the network). Therefore, in the application, the virtual side lane line of the second road section can be automatically generated according to the central position point on the central lane line of the second road section and the width information of the second road section, the process of generating the virtual side lane line is not needed to be participated manually, and the accuracy and the efficiency of generating the side lane line are improved. After the side virtual lane line of the road section is generated, the virtual side lane line is marked in the high-precision map comprising the second road section to obtain the marked high-precision map, and the marked fine map is sent to the vehicle which is about to run or runs on the second road section, so that the vehicle can realize navigation according to the marked high-precision map, the navigation accuracy is improved, and automatic driving or auxiliary driving is facilitated.

The above Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.

The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning, automatic driving, intelligent traffic and the like.

Machine Learning (ML) is a multi-domain cross discipline, and relates to a plurality of disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and the like. The special research on how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve the performance of the computer. Machine learning is the core of artificial intelligence, is the fundamental approach for computers to have intelligence, and is applied to all fields of artificial intelligence. Machine learning and deep learning generally include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and formal education learning.

In order to facilitate a clearer understanding of the present application, a lane line generating system for implementing the lane line generating method of the present application is first introduced, as shown in fig. 1, the lane line generating system includes a server 10 and a vehicle cluster, as shown in fig. 1, the vehicle cluster may include one or more vehicles, and the number of the vehicles is not limited herein. As shown in fig. 1, the vehicle cluster may specifically include vehicle 1, vehicle 2, …, vehicle n; it is understood that each vehicle has a terminal therein, which may be referred to as a vehicle-mounted terminal. As here the vehicle 1 may have a car terminal 1, the vehicle 2 may have a car terminal 2, … … and the vehicle n may have a car terminal n. As shown in fig. 1, the vehicle terminals in the vehicles 1, 2, 3, …, and n may be all connected to the server 10 via a network, so that the vehicle terminal in each vehicle may interact data with the server 10 via the network.

The server 10 may be a traffic management device, such as a road management device, and is configured to acquire a side lane line of the second road segment, mark the side lane line of the second road segment in a map including the second road segment, obtain a marked map, and send the marked map to a vehicle-mounted terminal of each vehicle. And the vehicle-mounted terminal of each vehicle is used for planning the driving route of the vehicle according to the marked map.

The server may be an independent physical server, a server cluster or a distributed system formed by at least two physical servers, or a cloud server providing basic cloud computing services such as cloud service, a cloud database, cloud computing, a cloud function, cloud storage, Network service, cloud communication, middleware service, domain name service, security service, Content Delivery Network (CDN), big data and an artificial intelligence platform. The vehicle-mounted terminal may specifically refer to a smart phone, a smart speaker, a sound box with a screen, a smart watch, and the like, but is not limited thereto. Each vehicle-mounted terminal and each server may be directly or indirectly connected through wired or wireless communication, and the number of the vehicle-mounted terminals and the number of the servers may be one or at least two, which is not limited herein.

The lane line generation method in the present application may be implemented based on the lane line generation system shown in fig. 1, and as shown in fig. 2, the lane line generation method may include the following steps s1 to s 4: s1, the server obtains the central position point on the central lane line of the second road section. As shown in fig. 2, the first road segment includes a first straight road segment and a second straight road segment having an adjacent relationship with the first straight road segment and the second straight road segment, respectively, the first straight road segment being located at an entrance position of the second road segment, and the second straight road segment being located at an exit position of the second road segment. The first straight road section and the second straight road section both include actual side lane lines, and as shown in fig. 2, the solid line portions at the first straight road section and the second straight road section are the actual side lane lines. The dotted line portion at the second road section is a center lane line of the second road section, and the server may randomly extract a plurality of center position points reflecting a contour of the center lane line of the second road section from the center lane line of the second road section. s2, encryption center point. After the server acquires the central position point, the distance between the adjacent position points is acquired, and if the distance between the adjacent position points is greater than a distance threshold value, the acquired central position point is excessively sparse, so that the virtual side lane line is easily excessively smooth, and therefore the central position point needs to be encrypted, namely more central position points are acquired from the central lane line of the second road section. s3, generating side position points. The server can determine the road section width of the second road section according to the road section width of the first straight road section and the road section width of the second straight road section, and directly translate the central position point according to the road section width of the second road section to obtain a side position point for reflecting the outline of the virtual side lane line of the second road section; alternatively, a direction vector is generated from the center position point, and a side position point for reflecting the contour of the virtual side lane line of the second road segment is obtained by performing translation, rotation, and the like on the direction vector. s4 smoothes the candidate side lane lines. After the server acquires the side position points, the server may sequentially connect the side position points according to the position information of the side position points to obtain candidate side lane lines. As shown in fig. 2, in the second link, the dotted line portion connected to the actual side lane lines of the first straight link and the second straight link is the virtual side lane line of the second link. Further, the server can mark the virtual side lane line in a map comprising a second road section to obtain a marked map, and send the marked map to a vehicle about to run or running on the second road section, so that the vehicle can realize navigation according to the marked map, and the navigation accuracy is improved.

Further, please refer to fig. 3, which is a schematic flow chart of a lane line generating method according to an embodiment of the present application. As shown in fig. 3, the method may be performed by the terminal in fig. 1, the server in fig. 1, or both the terminal and the server in fig. 1, and devices for performing the method in this application may be collectively referred to as computer devices. The lane line generation method comprises the following steps of S101-S105:

s101, acquiring an actual side lane line of a first road section and a center lane line of a second road section; the second road section and the first road section have adjacent relation, and the second road section does not contain an actual side lane line.

In the application, the computer device may obtain an actual side lane line of the first road segment and a center lane line of the second road segment from the image data including the first road segment and the second road segment; alternatively, the computer device may obtain the actual side lane lines of the first road segment and the center lane lines of the second road segment from a map including the first road segment and the second road segment. The second road section and the first road section have an adjacent relation, the first road section comprises an actual side lane line, and the second road section does not comprise the actual side lane line.

And S102, determining the width information of the second road section according to the actual side lane line of the first road section.

And S103, extracting a central position point reflecting the outline of the central lane line of the second road section from the central lane line of the second road section.

In steps S102-S103, the computer device may determine an exit width and an entrance width of a second road segment based on the actual side lane line of the first road segment, and determine width information of the second road segment based on the exit width and the entrance width of the second road segment. Then, the computer device may extract a center position point reflecting an outline of a center lane line of the second road segment from the center lane of the second road segment by the limit distance. The limit distance is a maximum distance, a minimum distance, or a maximum distance and a minimum distance of an allowable interval between center position points, and may be dynamically determined according to an application scenario, for example, the limit distance may be determined according to link attribute information of the second link, and the link attribute information of the second link includes information of a width, a length, an angle, and the like of the second link. Alternatively, the limit distance may be determined according to the user's needs. The central lane line of the second road segment may refer to a lane line located in the middle of the second road segment, that is, the central lane line of the second road segment is located in the middle of the virtual left and right lane lines of the second road segment, that is, the distance between the virtual left lane line of the second road segment and the central lane line of the second road segment is the same as the distance between the virtual right lane line of the second road segment and the central lane line of the second road segment; the contour of the center lane line of the second road segment may refer to an angle of the center lane line of the second road segment.

Specifically, the computer device may acquire the width information of the second road segment in any one of or a combination of the following two ways.

In the first mode, when the number of the first road segments is 1, for example, the first road segments are straight road segments located at the entrance position of the second road segments, or the first road segments are straight road segments located at the exit position of the second road segments. The computer device may determine a link width of the first link according to a distance between the virtual left lane line and the right lane line of the first link, regard the link width of the first link as a link width of the second link, and generate width information of the second link.

The second method comprises the following steps: when the number of the first road segments is 2, for example, the first road segments include a first straight road segment located at an entrance position of the second road segment and a second straight road segment located at an exit position of the second road segment. The computer device may determine a section width of the first straight section according to an actual side lane line of the first straight section, and determine a section width of the second straight section according to an actual side lane line of the second straight section; averaging the road section width of the first straight road section and the road section width of the second straight road section to obtain an average road section width; and generating width information of the second road section according to the average road section width.

In the second mode, the computer device may determine the link width of the first straight link according to a distance between the virtual left lane line and the right lane line of the first straight link, and determine the link width of the second straight link according to a distance between the virtual left lane line and the right lane line of the second straight link. Then, the average algorithm is adopted to average the link width of the first straight link and the link width of the second straight link to obtain an average link width, and the average link width is used as the link width of the second link to generate the width information of the second link. That is, by obtaining the link width of the second link by averaging the link width at the entrance and the link width at the exit of the second link, the accuracy of the link width of the second link can be improved.

For example, as shown in fig. 4, the first road segment includes a first straight road segment and a second straight road segment having an adjacent relationship with the first straight road segment and the second straight road segment, respectively, the first straight road segment being located at an entrance position of the second road segment, and the second straight road segment being located at an exit position of the second road segment. The first straight road section and the second straight road section both include actual side lane lines, and as shown in fig. 2, the solid line portions at the first straight road section and the second straight road section are the actual side lane lines. The computer device may take U, T from the actual left lane line and the actual right lane line of the first straight road segment, respectively, the connection line of U and T being perpendicular to the center lane line of the first straight road segment; and a position point R, S is taken from the actual left lane line and the actual right lane line of the first straight road section, and the connecting line of the position points R and S is vertical to the central lane line of the second straight road section. Further, the distance between the position points U and T can be obtained, and the distance between the position points U and T is used as the section width of the first straight section; the distance between the position points R and S may be acquired as the link width of the second straight link. Then, the link width of the first straight link and the link width of the second straight link may be averaged to obtain an average link width, which is used as the link width of the second link. For example, the link width of the second link may be calculated by the following formula (1):

wherein w of the formula (1) represents a link width of the second link,

road representing a first straight road sectionThe width of the segment is such that,

indicating the link width of the second straight road segment.

And S104, determining a side position point for reflecting the outline of the virtual side lane line of the second road section according to the center position point and the width information of the second road section.

And S105, generating a virtual side lane line of the second road section according to the side position point.

In steps S104 to S105, since there is a similarity between the virtual side lane line of the second road segment and the center lane line of the second road segment, that is, there is a similarity between the outline of the virtual side lane line of the second road segment and the outline of the center lane line of the second road segment, the computer device may determine the virtual side lane line of the second road segment according to the center position point and the width information of the second road segment. Specifically, the computer device may determine, according to the center position point and the width information of the second road segment, a side position point for reflecting an outline of the virtual side lane line of the second road segment; here, the side position points include a left position point and a right position point, the left position point is a contour for reflecting a virtual left lane line of the second link, and the right position point is a contour for reflecting a virtual right lane line of the second link. Further, the computer device may generate a virtual left lane line for the second road segment from the left location point and a virtual right lane line for the second road segment from the right location point.

Optionally, the computer device here is a centralized management device of a road, the centralized management device acquires a map including the second road segment after marking the virtual side lane line of the second road segment in the map, and obtains a marked map, and then, the marked map may be sent to a vehicle driving in the second road segment; so that the vehicle can realize navigation according to the marked map, and the navigation accuracy is improved; meanwhile, the marked map is distributed to each vehicle by the centralized management equipment, and each vehicle is not required to independently obtain the virtual lane line of the second road section, so that the resource waste is avoided, and the driving speed of the vehicle is improved.

For convenience of distinction, the computer device may mark the virtual side lane line of the second road segment in the map according to a target marking manner, to obtain a marked map, where the target marking manner may be a marking manner different from a marking manner of the actual side lane line of the first road segment, and the target marking manner includes a lane line color, a lane line thickness, a lane line type (such as a solid line or a dotted line), and the like.

Further, please refer to fig. 5, which is a schematic flow chart of a lane line generating method according to an embodiment of the present application. As shown in fig. 5, the method may be performed by the terminal in fig. 1, the server in fig. 1, or both the terminal and the server in fig. 1, and devices for performing the method in this application may be collectively referred to as computer devices. The lane line generation method may include the following steps S201 to S207:

s201, acquiring an actual side lane line of a first road section and a center lane line of a second road section; the second road section and the first road section have adjacent relation, and the second road section does not contain an actual side lane line.

S202, determining width information of the second road section according to the actual side lane line of the first road section.

In this application, the actual side lane line of first highway section includes actual left side lane line and actual right side lane line, can acquire the distance between actual left side lane line and the actual right side lane line, determines the highway section width of first highway section with the distance between actual left side lane line and the actual right side lane line. Since the link widths between the links having the adjacent relationship are equal or approximately equal, the computer device may determine the link width of the second link from the link width of the first link, and generate the width information of the second link from the link width of the second link.

S203, extracting a center position point reflecting the contour of the center lane line of the second road segment from the center lane line of the second road segment.

Optionally, if the center position point extracted from the center lane line of the second road segment is excessively sparse, the side position point may be excessively sparse, and further, the virtual side lane line may be easily caused to be excessively smooth, even if the virtual side lane line is distorted; if the center position points extracted from the center lane line of the second road section are excessively dense, the side position points are excessively dense, and further, failure in acquiring the virtual side lane line is easily caused, and excessive resources are consumed. Therefore, the computer device may extract the center position point from the center lane line of the second road segment based on the link attribute information of the second road segment, and in particular, the computer device may acquire the link attribute information of the second road segment, where the link attribute information of the second road segment includes one or more of a width of the second road segment, a length of the second road segment (i.e., a length of the center lane line), and an angle, the second road segment is typically a turning link, and the angle of the second road segment may be referred to as a turning angle of the second road segment. Further, a limit distance between the center position points may be determined according to the link attribute information of the second link, the limit distance being a maximum distance, a minimum distance, or a maximum distance and a minimum distance of an allowable interval between the center position points. Then, extracting a center position point for reflecting the contour of the center lane line from the center lane line of the second road section according to the limit distance; the distance between adjacent ones of the center position points is smaller than the limit distance. By extracting the central position point from the central lane line of the second road section according to the limiting distance, the acquired central position point can be effectively prevented from being excessively sparse or excessively dense, the accuracy of acquiring the central position point is improved, and the problem of resource waste caused by the fact that the central position point is fixed and dense is solved.

It should be noted that, when the limitation distance includes the maximum distance, the distance between center position points having an adjacent relationship among the center position points is smaller than the maximum distance; when the restriction distance includes the minimum distance, a distance between center position points having an adjacent relationship among the center position points is larger than the minimum distance; when the limit distance includes a maximum distance and a minimum distance, a distance between center position points having an adjacent relationship among the center position points is smaller than the maximum distance and larger than the minimum distance.

It should be noted that the limiting distance may be obtained by identifying the link attribute information of the second link by using a target distance obtaining model, where the target distance obtaining model is obtained by training a candidate distance obtaining model by using link attribute information of a large number of sample links and a labeled limiting distance, and the labeled limiting distance may be obtained by manually analyzing the link attribute information of the sample links to obtain an optimal limiting distance. The generation process of the target distance acquisition model comprises the following steps: the computer device can obtain the road section attribute information and the marked limit distance of the sample road section, and predict the road section attribute information of the sample road section by adopting the candidate distance obtaining model to obtain the predicted limit distance. And determining a distance prediction error of the candidate distance acquisition model according to the distance limit distance and the labeling limit distance, wherein if the distance prediction error is in a convergence state, the prediction limit distance is similar to or the same as the labeling limit distance, namely the prediction accuracy of the candidate distance acquisition model is higher, and the candidate distance acquisition model can be used as a target distance acquisition model. If the distance prediction error is not in a convergence state, the difference between the prediction limit distance and the labeling limit distance is large, namely the prediction accuracy of the candidate distance acquisition model is low, and the candidate distance acquisition model can be adjusted according to the distance prediction error to obtain the target distance acquisition model.

And S204, determining a side position point for reflecting the outline of the virtual side lane line of the second road section according to the center position point and the width information of the second road section.

Optionally, the center position points include a first center position point, a second center position point, and a third center position point; the first central position point and the second central position point have an adjacent relation, and the second central position point and the third central position point have an adjacent relation; the determining a side position point for reflecting the outline of the virtual side lane line of the second road section according to the center position point and the width information of the second road section includes: acquiring a first direction vector for reflecting the direction from the first central position point to the second central position point, acquiring a second direction vector for reflecting the direction from the second central position point to the third central position point, and averaging the first direction vector and the second direction vector to obtain a first average vector; and determining a side position point for reflecting the outline of the virtual side lane line of the second road section according to the first average vector and the width information of the second road section.

For example, the first direction vector may refer to a unit direction vector from the first center position point to the second center position point, and the second direction vector may refer to a unit direction vector from the second center position point to the third center position pointUnit direction vector of the point. The first center position point, the second center position point and the third center position point are respectively represented by letters A, B, C, and the computer device can be used for vector

And the distance from the point A to the point B

The computer device may use the ratio of the first direction vector to the second direction vector

And the distance from the point B to the point C

The ratio therebetween is used as the second direction vector, and then, the first direction vector and the second direction vector may be averaged to obtain the first average vector. The first direction vector, the second direction vector, and the first average vector may be expressed by the following equation (2):

wherein the content of the first and second substances,

the method includes respectively representing a first direction vector, a second direction vector and a first average vector, after the first average vector is obtained, performing operations such as rotation and translation on the first average vector to obtain a side position point for reflecting the outline of the virtual side lane line of the second road segment, and specifically includes: acquiring a virtual side lane line to be determined in the side direction of the second road section, and translating the first average vector to the second central position point; rotating the first average vector to the side direction by a specified angle at the second central position point to obtain a second average vector; the specified angle is based on the second road segmentIs determined by the contour of the central lane line of (1); and translating the second average vector to the side direction according to the width information of the second road section to obtain a side position point for reflecting the outline of the virtual side lane line of the second road section.

For example, when it is necessary to determine the virtual left lane line of the second road segment, the computer device may translate the first average vector to a second center position point with the left direction as a side direction, rotate the first average vector at the second center position point by a specified angle to the left direction, and obtain a second average vector. And translating the second average vector to the left direction by a specified distance to obtain a left position point for reflecting the outline of the virtual left lane line of the second road section, and repeatedly executing the steps to obtain a plurality of left position points for reflecting the outline of the virtual left lane line of the second road section. Similarly, when the virtual right lane line of the second road section needs to be determined, the computer device may use the right direction as the side direction, translate the first average vector to the second center position point, and rotate the first average vector at the second center position point by a specified angle in the right direction to obtain the second average vector. And translating the second average vector to a specified distance in the right direction to obtain a right position point for reflecting the outline of the virtual right lane line of the second road section, and repeatedly executing the steps to obtain a plurality of right position points for reflecting the outline of the virtual right lane line of the second road section.

For example, as shown in fig. 6, the center position point A, B, C, D is included on the center lane line of the second road segment, and the computer device may sequentially traverse three center position points having an adjacent relationship on the center lane line of the second road segment, first, traverse the center position point A, B, C, and obtain a unit vector of the center position point a to the center position point B

Obtaining the unit vector from the center position point B to the center position point C

Then, a pair unit vector is calculated

Sum unit vector

Carrying out averaging processing to obtain a first average vector

And moving the first average vector to a central position point B, rotating the first unit vector at the central position point B by 90 degrees towards the left direction, and translating the first unit vector leftwards by a distance of w/2 to obtain a left position point F. Similarly, the first unit vector is rotated by 90 degrees to the right direction at the central position point B, and the first unit vector is translated leftward by a distance of w/2, so as to obtain a right position point J. Further, the computer device may traverse center position point B, C, D, repeating the above steps, resulting in left position point J and right position point K. Specifically, the computer device may translate the central position point a leftward by a distance of w/2 to obtain a left position point E, and translate the central position point a rightward by a distance of w/2 to obtain a right position point I; similarly, the central position point D is translated to the left by a distance of w/2 to obtain a left position point H, and the central position point D is translated to the right by a distance of w/2 to obtain a right position point L. As in fig. 6, the left position point E, F, G, H of the contour reflecting the virtual left lane line of the second link is obtained by the above steps; and a right position point I, J, K, L may be obtained that reflects the outline of the virtual right lane line for the second road segment.

S205, acquiring the position information of at least two side position points.

S206, sequentially connecting the at least two side position points according to the position information of the at least two side position points to obtain candidate side lane lines.

S207, smoothing the candidate side lane line to obtain a virtual side lane line of the second road section.

In steps S205 to S207, the side position points include a left position point and a right position point, the number of the left position point and the right position point is at least two, the computer device may obtain position information of each left position point in the at least two left position points, and sequentially connect the left position points according to the position information of each left position point, that is, connect left position points having an adjacent relationship to each other, to obtain a candidate left lane line, and smooth-process the candidate left lane line, to obtain a virtual left lane line of the second road section. Similarly, the computer device may obtain position information of each of the at least two right position points, sequentially connect the right position points according to the position information of the right position points, to connect right position points that are to have an adjacent relationship, to obtain a candidate right lane line, and perform smoothing processing on the candidate right lane line to obtain a virtual right lane line of the second road segment. By smoothing the candidate left lane line and the candidate right lane line, the smoothness of the virtual lane line of the second road section is improved, so that the virtual side lane line of the second road section is closer to the actual side lane line, and the accuracy of obtaining the side lane line is improved.

Optionally, the computer device may obtain a smoothing parameter related to the candidate side lane line through the target parameter obtaining model, and perform smoothing on the candidate side lane line according to the smoothing parameter to obtain a virtual side lane line of the second road segment. Specifically, a target parameter acquisition model is adopted to identify the position information of the at least two side position points, and a smoothing processing parameter related to the candidate side lane line is obtained; and smoothing the candidate side lane line according to the smoothing parameter to obtain a virtual side lane line of the second road section.

For example, the candidate side lane lines include a candidate left lane line and a candidate right lane line, and for the candidate left lane line, the computer device may identify the position information of the at least two left position points by using the target parameter acquisition model to obtain a first smoothing parameter about the candidate left lane line; and smoothing the candidate left lane line according to the first smoothing parameter to obtain a virtual left lane line of the second road section. Similarly, the computer device may identify the position information of the at least two right position points by using a target parameter obtaining model to obtain a second smoothing processing parameter related to the candidate right lane line; and smoothing the candidate right lane line according to a second smoothing parameter to obtain a virtual right lane line of the second road section. The smoothing processing parameters are acquired by adopting a machine learning technology in artificial intelligence, so that the accuracy of the smoothing processing parameters is improved, the problem that the smoothness of the virtual side lane line of the second road section is poor is solved, and the problem that the virtual side lane line of the second road section is excessively smooth is solved.

Wherein, the smoothing processing parameter includes any one parameter or two parameters of the limiting angle between the lane line segments, the iteration limiting times of the smoothing processing.

Optionally, when the smoothing processing parameter includes a limit angle between lane line segments; the smoothing of the candidate side lane line according to the smoothing parameter to obtain the virtual side lane line of the second road segment includes: sequentially traversing three side position points with adjacent relation on the candidate side lane line to obtain a first side lane line segment and a second side lane line segment on the candidate side lane line; the first side lane line segment is formed by a first side position point P_i-2And a second side position point P_i-1The second side lane line segment is composed of a second side position point P_i-1And a third side position point P_iForming; i is an integer greater than 3 and less than or equal to M, M being the number of lateral position points on the candidate lateral lane line. Determining a target angle between an extension lane line segment of the first side lane line segment and the second side lane line segment; and if the target angle is larger than the limiting angle, carrying out angle adjustment on the candidate side lane line according to the smoothing processing parameter to obtain a virtual side lane line of the second road section.

For example, traversing the candidate left lane line in sequence with any neighborsThree lateral position points of relation, connecting the first left position point P_i-2And said second left position point P_i-1Obtaining a first left lane line segment, connecting a second left position point P_i-1And said third left position point P_iAnd obtaining a second left lane segment. A first target angle between an extension lane segment of the first left lane segment and the second left lane segment is determined. If the first target angle is less than or equal to the limit angle, it indicates that the smoothness of the candidate left lane line is relatively high, and therefore, the candidate left lane line may be used as the virtual left lane line of the second link. If the first target angle is larger than the limiting angle, it is indicated that smoothness of the candidate left lane line is low, and therefore, angle adjustment can be performed on the candidate left lane line according to the first smoothness processing parameter to obtain a virtual left lane line of the second road section. Similarly, sequentially traversing any three side position points with adjacent relation on the candidate left lane line, and connecting the first right position point P_i-2And said second right position point P_i-1Obtaining a first right lane line segment, connecting the second right position point P_i-2And said third right position point P_iAnd obtaining a second right lane segment. Determining a second target angle between an extension lane segment of the first right lane segment and the second right lane segment. If the second target angle is less than or equal to the limit angle, it indicates that the smoothness of the candidate right lane line is relatively high, and therefore, the candidate right lane line may be used as the virtual right lane line of the second link. If the second target angle is larger than the limiting angle, it is indicated that smoothness of the candidate right lane line is low, and therefore, angle adjustment can be performed on the candidate right lane line according to the second smoothness processing parameter, so that a virtual right lane line of the second road section is obtained. Whether to carry out smoothing processing on the candidate side lane line according to the smoothing processing parameters is determined according to any three side position points with adjacent relations on the side lane line, so that the accuracy of carrying out smoothing processing on the candidate lane line is improved, and the virtual side lane line can be effectively prevented from being excessiveSmoothness, or less smoothness.

It should be noted that, the process of performing the smoothing process on the candidate side lane lines refers to: the three side position points with adjacent relation on the candidate side lane line are smoothed in sequence, and the smoothed side position points are affected in the smoothing process of the rear side position points, so that the candidate side lane line needs to be subjected to multiple smoothing iterations to achieve the overall smoothing effect. One iteration of the smoothing process for the candidate side lane lines is as follows: at least one traversal operation is completed on any three side position points with adjacent relation on the candidate side lane line; if the iteration times are too large, the problem of excessive smoothness of the candidate side lane lines is easily caused. Meanwhile, the problem of excessive smoothness of candidate side lane lines is easily caused by the fact that the angles between lane line segments are too small. Therefore, the computer device may control the timing of ending the smoothing of the candidate side lane line according to any one or both of the limiting angle and the iteration limiting number in the smoothing parameters. For example, the computer device may adopt any one of three control manners or a plurality of combination control manners to control the timing of ending the smoothing processing on the candidate side lane line.

The control method is as follows: the timing of ending the smoothing process on the candidate side lane lines is controlled by the angle limit between the lane line segments. Specifically, if the target angle is greater than the limiting angle, it indicates that the angle between the first side lane line segment and the second side lane line segment is not smooth enough, and therefore, the first side position point P can be obtained_i-2And said third side position point P_iAnd forming a third side lane segment. Then, the second side position point P is set_i-1Translating the lane line segment to the direction close to the third side to obtain a fourth side position point Q_i-1Obtaining the first side position point P_i-2And said fourth side position point Q_i-1A fourth lateral lane segment formed, and a fourth lateral positionPoint Q_i-1And said third side position point P_iA fifth lateral lane segment; if the angle between the extension line of the fourth side lane line segment and the fifth side lane line segment is smaller than or equal to the limit angle, that is, for any three side position points on the candidate side lane line having an adjacent relationship, when the angle between the corresponding extension lane line segment and the lane line segment is smaller than or equal to the limit angle, finishing the smoothing processing of the candidate side lane line, and generating the virtual side lane line of the second road section according to the fourth side lane line segment and the fifth side lane line segment. The time for finishing the smoothing treatment of the candidate side lane lines is controlled through the limiting angle between the lane line segments, so that the virtual side lane lines of the second road section can achieve the effect of integral smoothing, the smoothing effect of the virtual side lane lines of the second road section is improved, and the virtual side lane lines of the second road section are closer to the actual side lane lines.

And a second control mode: and controlling the time for finishing the smoothing processing on the candidate side lane lines by the iteration limit times among the lane line segments. Specifically, if the target angle is greater than the limiting angle, it indicates that the angle between the first side lane line segment and the second side lane line segment is not smooth enough, and therefore, the first side position point P can be obtained_i-2And said third side position point P_iAnd forming a third side lane segment. Then, the second side position point P is set_i-1Translating the lane line segment to the direction close to the third side to obtain a fourth side position point Q_i-1Obtaining the first side position point P_i-2And said fourth side position point Q_i-1A fourth side lane line segment formed, and a fourth side position point Q_i-1And said third side position point P_iAnd forming a fifth side lane segment. When any three side position points with adjacent relation on the candidate side lane line finish traversing operation, counting iteration times of smoothing processing; if the iteration times are greater than the iteration limit times, the fourth side lane line segment and the fifth side lane line segment are determinedAnd generating a virtual side lane line of the second road section. That is, when the number of traversal operations for any side position point on the candidate side lane line is greater than the number of iteration limits, the smoothing processing on the candidate side lane line is finished, and the virtual side lane line of the second road segment is generated according to the fourth side lane line segment and the fifth side lane line segment. The time for finishing the smoothing treatment of the candidate side lane line is controlled by the iteration limit times of the smoothing treatment, so that the virtual side lane line of the second road section can achieve the effect of integral smoothing, the smoothing effect of the virtual side lane line of the second road section is improved, and the virtual side lane line of the second road section is closer to the actual side lane line.

And a third control mode: as shown in fig. 7, the timing of ending the smoothing process for the side-candidate lane lines is controlled by the number of iterative restrictions between lane line segments and the restriction angle between lane line segments. Specifically, the method comprises the following steps s 11-s 15:

s11, if the target angle is larger than the limiting angle, obtaining the first side position point P_i-2And said third side position point P_iAnd forming a third side lane segment.

s12, connecting the second side position point P_i-1Translating the lane line segment to the direction close to the third side to obtain a fourth side position point Q_i-1。

s13, obtaining the first side position point P_i-2And said fourth side position point Q_i-1A fourth side lane line segment formed, and a fourth side position point Q_i-1And said third side position point P_iAnd forming a fifth side lane segment.

s14, when any three side position points with adjacent relation on the candidate side lane line finish the traversing operation, counting the iteration times of the smoothing processing.

s15, if the angle between the extension line of the fourth side lane line segment and the fifth side lane line segment is less than or equal to the limit angle, or the iteration number is greater than the iteration limit number, generating the virtual side lane line of the second road segment according to the fourth side lane line segment and the fifth side lane line segment.

In steps s 11-s 15, if the first angle is larger than the limiting angle, it indicates that the angle between the first side lane segment and the second side lane segment is not smooth enough, so that the first side position point P can be obtained_i-2And the third side position point P_iAnd forming a third side lane segment. Then, the second side position point P is set_i-1Translating the lane line segment to the direction close to the third side to obtain a fourth side position point Q_i-1Obtaining the first side position point P_i-2And said fourth side position point Q_i-1A fourth side lane line segment formed, and a fourth side position point Q_i-1And said third side position point P_iAnd forming a fifth side lane segment. When any three side position points with adjacent relation on the candidate side lane line finish traversing operation, counting iteration times of smoothing processing; if the angle between the extension of the fourth side lane segment and the fifth side lane segment is less than or equal to the limit angle, or the iteration number is greater than the iteration limit number. That is, when the number of traversal operations for any side position point on the candidate side lane line is greater than the iteration limit number, or for any three side position points having an adjacent relationship on the candidate side lane line, when the corresponding angles between the extension lane line segment and the lane line segment are less than or equal to the limit angle, the smoothing of the candidate side lane line is finished, and then the virtual side lane line of the second road segment is generated according to the fourth side lane line segment and the fifth side lane line segment. The time for finishing the smoothing treatment of the candidate side lane line is controlled through the iteration limit times of the smoothing treatment and the limit angle between lane line segments, so that the virtual side lane line of the second road section can achieve the effect of integral smoothing, the smoothing effect of the virtual side lane line of the second road section is improved, and the virtual side lane line of the second road section is closer to the real side lane lineA boundary side lane line.

Optionally, the translating the second side position point to a direction close to the third side lane line segment to obtain a fourth side position point includes: the computer device can determine a center point on the third side lane segment from the center point on the third side lane segment to the first side point P_i-2From the center point on the third side lane line segment to the third side position point P_iAre the same. Further, the computer device may obtain a sixth side lane segment composed of the center position point on the third side lane segment and the second side position point, determine the center position point on the sixth side lane segment, and determine the second side position point P_i-1Translating to the central position point on the sixth side lane line segment to obtain the fourth side position point Q_i-1. The angle between the extension lane line segment and the lane line segment is reduced by moving any side position point in the middle position among the three side position points with adjacent relation, so that the candidate side lane line is subjected to smoothing treatment, and resources can be saved.

For example, as shown in fig. 8, the left position point of the outline for reflecting the virtual left lane line of the second road segment includes E, F, G, H; the right position point of the outline of the virtual right lane line for reflecting the second road segment includes I, J, K, L. The left position point E, F, G, H is sequentially connected to obtain a left lane line candidate for the second link, and the right position point I, J, K, L is sequentially connected to obtain a right lane line candidate for the second link. First, the computer device may smooth the candidate left lane line of the second road segment, and specifically, the computer device traverses the left position point E, F, G to obtain an angle between the extended lane line segment of the lane line segment EF and the lane line segment FG, if the angle between the extended lane line segment of the lane line segment EF and the lane line segment FG is greater than the limit angle, connects the left position point E, G to obtain the lane line segment EG, and connects the center position point M of the lane line segment EG to the left position point M to obtain the lane line segment MF, and moves the left position point F to the center position point N of the lane line segment MF. If the angle between the extension lane segment of lane segment EN and lane segment NG is less than the limit angle, a virtual left lane line for the second road segment formed by lane segment EN and lane segment NG may be generated. If the angle between the extension lane segment of the lane segment EN and the lane segment NG is larger than the limit angle, the central position point N is moved to the central position of the lane segment MN, the steps are repeatedly executed until the angle between the extension lane segment of the lane segment and another lane segment is smaller than or equal to the limit angle, then, the steps are traversed F, G, H, the iteration number is increased by 1, and whether the iteration number is larger than the iteration limit number is judged. If the iteration number is greater than the iteration limit number, the smoothing processing on the candidate left lane line is finished, and if the iteration number is less than or equal to the iteration limit number, the left position point E, F, G may be traversed again, and the above steps are repeated to obtain the virtual left lane line of the second road segment. Similarly, the computer device may sequentially traverse each right position point, and perform smoothing processing on the candidate right lane line to obtain a virtual right lane line of the second road segment, where the specific execution process may refer to the execution process for obtaining the virtual left lane line of the second road segment, and repeated parts are not described again.

It should be noted that, if the center position points extracted from the center lane line of the second road segment are sparse, the side position points obtained by calculation are also sparse, and the shape of the original lane line is more easily changed when the candidate side lane line is subjected to smoothing processing, which results in distortion of the obtained virtual lane line. For example, as shown in fig. 9, if the center position points extracted from the center lane line of the second link are sparse, it is easy to cause the virtual left-side lane line and the virtual right-side lane line to be excessively smooth and to have a large difference from the contour of the center lane line. In order to ensure that the outline of the virtual side lane does not change excessively, the density of central position points on the central lane line is controlled before the side position points are calculated, and if the density is too low, encryption is carried out, namely the number of the central position points is increased; if the density is too high, thinning is performed, i.e., the number of center points is reduced. As shown in fig. 9, when the density of the center position point on the center lane is too low, the center position point is encrypted, candidate side lane lines are determined according to the encrypted center position point, and the candidate side lane lines are smoothed, so that the virtual side lane lines of the second road segment are obtained. As shown in fig. 9, the virtual left lane line and the virtual right lane line obtained by encrypting the center point have profiles similar to the profile of the center lane line, that is, the profile of the virtual side lane line is not changed, so that the accuracy of obtaining the virtual side lane line is improved. Wherein the density of the center position points on the center lane line may be determined according to a distance between adjacent center position points.

Optionally, the computer device may train the candidate parameter obtaining model by using a large number of sample side position points on the sample side lane line and the labeled smoothing processing parameters of the sample lane line, so as to obtain the target parameter obtaining model, which is beneficial to improving the accuracy of the parameters obtained by the target parameter obtaining model. Specifically, the computer device may obtain position information of a sample side position point for reflecting a contour of a sample side lane line and a smoothing parameter for labeling with respect to the sample side lane line, and predict the position information of the sample side position point using a candidate parameter obtaining model to obtain a smoothing parameter for predicting with respect to the sample side lane line. And adjusting the candidate parameter acquisition model according to the labeled smoothing processing parameter and the predicted smoothing processing parameter to obtain the target parameter acquisition model.

The computer device may obtain position information of a sample side position point for reflecting the contour of the sample side lane line, and a smoothing parameter related to the labeling of the sample side lane line, where the smoothing parameter may be obtained by a professional analyzing the position information of the sample side position point of the contour of the sample side lane line. Then, predicting the position information of the sample side position point by adopting a candidate parameter acquisition model to obtain a predicted smoothing processing parameter related to the sample side lane line, wherein if the difference between the predicted smoothing processing parameter and the labeled smoothing processing parameter is small, the accuracy of the candidate parameter acquisition model is high; if the difference between the predicted smoothing processing parameter and the labeled smoothing processing parameter is larger, the accuracy of the candidate parameter obtaining model is lower. Therefore, the computer device can adjust the candidate parameter acquisition model according to the labeled smoothing processing parameter and the predicted smoothing processing parameter to obtain the target parameter acquisition model; the candidate parameter acquisition model is trained by adopting a large number of sample side position points on the sample side lane line and the labeled smoothing processing parameters of the sample lane line to obtain a target parameter acquisition model, so that the accuracy of the parameters acquired by the target parameter acquisition model is improved.

Optionally, the adjusting the candidate parameter obtaining model according to the labeled smoothing processing parameter and the predicted smoothing processing parameter to obtain the target parameter obtaining model includes: determining a parameter prediction error of the candidate parameter acquisition model according to the labeled smoothing processing parameter and the predicted smoothing processing parameter, and if the parameter prediction error is not in a convergence state, adjusting the candidate parameter acquisition model according to the parameter prediction error to obtain an adjusted candidate parameter acquisition model; and determining the adjusted candidate parameter acquisition model as the target parameter acquisition model.

The computer device may determine a parameter prediction error of the candidate parameter acquisition model according to the labeled smoothing processing parameter and the predicted smoothing processing parameter, and if the parameter prediction error is in a convergence state, that is, the parameter prediction error is smaller than a prediction error threshold, it indicates that the accuracy of the parameter acquired by the candidate parameter acquisition model is higher, and the candidate parameter acquisition model may be directly used as the target parameter acquisition model. If the parameter prediction error is not in a convergence state, that is, the parameter prediction error is greater than or equal to the prediction error threshold, it indicates that the accuracy of the parameter acquired by the candidate parameter acquisition model is relatively low. Therefore, the candidate parameter acquisition model is adjusted according to the parameter prediction error to obtain an adjusted candidate parameter acquisition model; and determining the adjusted candidate parameter acquisition model as the target parameter acquisition model. The candidate parameter acquisition model is trained according to the parameter prediction error of the candidate parameter acquisition model, so that the parameter acquisition error of the target parameter acquisition model is favorably reduced, and the accuracy of the parameters acquired by the target parameter acquisition model is improved.

In the application, the computer device may obtain an actual side lane line of the first road segment and a center lane line of the second road segment, where the first road segment and the second road segment have an adjacent relationship, and the second road segment does not include the actual side lane line, that is, the second road segment does not lack the side lane line; then, width information of the second road section is determined by the actual side lane line of the first road section, and a center position point reflecting the contour of the center line of the second road section is extracted from the center lane line of the second road section. Because the side lane line of the second road section has similarity with the center lane line of the second road section, that is, the profile of the side lane line of the second road section has similarity with the profile of the center lane line of the second road section; therefore, the side position points for reflecting the virtual side lane lines of the second road section can be determined by the center position point and the width of the second road section, the side position points are sequentially connected based on the position information of the side position points to obtain candidate side lane lines, and the candidate side lane lines are smoothed to obtain the virtual side lane lines of the second road section. Therefore, in the application, the virtual side lane line of the second road section can be automatically generated according to the central position point on the central lane line of the second road section and the width information of the second road section, the process of generating the virtual side lane line is not needed to be participated manually, and the accuracy and the efficiency of generating the side lane line are improved. Meanwhile, the candidate side lane lines are subjected to smoothing processing, so that the smoothness of the virtual lane lines of the second road section is improved, the virtual side lane lines of the second road section are closer to the actual side lane lines, and the accuracy of obtaining the side lane lines is improved. The scheme can be applied to scenes such as automatic driving, auxiliary driving and the like, and accurate navigation information can be provided for the vehicle according to the virtual side lane line.

Fig. 10 is a schematic structural diagram of a lane line generating device according to an embodiment of the present application. The lane line generating device may be a computer program (including program code) running on a computer device, for example, the lane line generating device is an application software; the apparatus may be used to perform the corresponding steps in the methods provided by the embodiments of the present application. As shown in fig. 10, the lane line generation apparatus may include: the device comprises an acquisition module 111, a determination module 112, an extraction module 113, a generation module 114 and an adjustment module 115.

Optionally, the center position points include a first center position point, a second center position point, and a third center position point; the first central position point and the second central position point have an adjacent relation, and the second central position point and the third central position point have an adjacent relation;

the determining module determines a side position point for reflecting the outline of the virtual side lane line of the second road section according to the center position point and the width information of the second road section, including:

acquiring a first direction vector for reflecting a direction from the first center position point to the second center position point, and acquiring a second direction vector for reflecting a direction from the second center position point to the third center position point;

averaging the first direction vector and the second direction vector to obtain a first average vector;

and determining a side position point for reflecting the outline of the virtual side lane line of the second road section according to the first average vector and the width information of the second road section.

Optionally, the determining module determines, according to the first average vector and the width information of the second road segment, a side position point for reflecting an outline of a virtual side lane line of the second road segment, including:

acquiring a virtual side lane line to be determined in the side direction of the second road section;

translating the first average vector to the second center location point;

rotating the first average vector to the side direction by a specified angle at the second central position point to obtain a second average vector; the specified angle is determined according to the outline of the central lane line of the second road section;

and translating the second average vector to the side direction according to the width information of the second road section to obtain a side position point for reflecting the outline of the virtual side lane line of the second road section.

Optionally, the number of the side position points is at least two; the generating module generates a virtual side lane line of the second road section according to the side position point, including:

acquiring position information of at least two side position points;

sequentially connecting the at least two side position points according to the position information of the at least two side position points to obtain candidate side lane lines;

and smoothing the candidate side lane line to obtain a virtual side lane line of the second road section.

Optionally, the generating module performs smoothing processing on the candidate side lane line to obtain a virtual side lane line of the second road segment, including:

adopting a target parameter acquisition model to identify the position information of the at least two side position points to obtain a smoothing processing parameter related to the candidate side lane line;

and smoothing the candidate side lane line according to the smoothing parameter to obtain a virtual side lane line of the second road section.

Optionally, the smoothing processing parameter includes a limit angle between lane line segments; the generating module performs smoothing processing on the candidate side lane line according to the smoothing processing parameter to obtain a virtual side lane line of the second road section, and the generating module includes:

sequentially traversing three side position points with adjacent relation on the candidate side lane line to obtain a first side lane line segment and a second side lane line segment on the candidate side lane line; the first side lane line segment is formed by a first side position point P_i-2And a second side position point P_i-1The second side lane line segment is composed of a second side position point P_i-1And a third side position point P_iForming; i is an integer greater than 3 and less than or equal to M, M being the number of side position points on the candidate side lane line;

determining a target angle between an extension lane line segment of the first side lane line segment and the second side lane line segment;

and if the target angle is larger than the limiting angle, carrying out angle adjustment on the candidate side lane line according to the smoothing processing parameter to obtain a virtual side lane line of the second road section.

Optionally, the smoothing parameter further includes an iteration limit number of smoothing; if the target angle is greater than the limiting angle, the generating module adjusts the angle of the candidate side lane line according to the smoothing parameter to obtain a virtual side lane line of the second road section, including:

if the target angle is larger than the limiting angle, acquiring the position point P of the first side edge_i-2And said third side position point P_iA third side lane segment;

the second side position point P_i-1Translating the lane line segment to the direction close to the third side to obtain a fourth side position point Q_i-1；

Obtaining the first side position point P_i-2And said fourth side position point Q_i-1A fourth side lane line segment formed, and a fourth side position point Q_i-1And said third side position point P_iA fifth lateral lane segment;

when any three side position points with adjacent relation on the candidate side lane line finish the traversal operation, counting the iteration times of the smoothing processing;

and if the angle between the extension line of the fourth side lane line segment and the fifth side lane line segment is smaller than or equal to the limit angle, or the iteration times are greater than the iteration limit times, generating the virtual side lane line of the second road section according to the fourth side lane line segment and the fifth side lane line segment.

Optionally, the generating module is configured to determine the second side position point P_i-1Translating the lane line segment to the direction close to the third side to obtain a fourth side position point Q_i-1The method comprises the following steps:

determining a center position point on the third side lane line segment, and acquiring a sixth side lane line segment formed by the center position point on the third side lane line segment and the second side position point;

determining a center position point on the sixth lateral lane segment;

the second side position point P_i-1Translating to the central position point on the sixth side lane line segment to obtain the fourth side position point Q_i-1。

Optionally, the apparatus further comprises an adjusting module 115, and the adjusting module 115 is configured to:

acquiring position information of a sample side position point for reflecting the outline of a sample side lane line and a label smoothing processing parameter related to the sample side lane line;

predicting the position information of the sample side position point by adopting a candidate parameter acquisition model to obtain a predicted smoothing processing parameter related to the sample side lane line;

and adjusting the candidate parameter acquisition model according to the labeled smoothing processing parameter and the predicted smoothing processing parameter to obtain the target parameter acquisition model.

Optionally, the adjusting module adjusts the candidate parameter obtaining model according to the labeled smoothing processing parameter and the predicted smoothing processing parameter to obtain the target parameter obtaining model, and the adjusting module includes:

determining a parameter prediction error of the candidate parameter acquisition model according to the labeled smoothing processing parameter and the predicted smoothing processing parameter;

if the parameter prediction error is not in a convergence state, adjusting the candidate parameter acquisition model according to the parameter prediction error to obtain an adjusted candidate parameter acquisition model;

and determining the adjusted candidate parameter acquisition model as the target parameter acquisition model.

Optionally, the extracting module extracts a center position point for reflecting the contour of the center lane line of the second road segment from the center lane line of the second road segment, and includes:

acquiring road section attribute information of the second road section;

determining a limiting distance between the central position points according to the road section attribute information of the second road section;

and extracting a center position point reflecting the contour of the center lane line from the center lane line of the second road section according to the limit distance.

Optionally, the first road segment includes a first straight road segment and a second straight road segment; the determining module determines the width information of the second road section according to the actual side lane line of the first road section, and the determining module comprises:

determining the section width of the first straight road section according to the actual side lane line of the first straight road section, and determining the section width of the second straight road section according to the actual side lane line of the second straight road section;

averaging the road section width of the first straight road section and the road section width of the second straight road section to obtain an average road section width;

and generating width information of the second road section according to the average road section width.

Optionally, the obtaining module is further configured to obtain a map including the second road segment; marking the virtual side lane line of the second road section in the map to obtain a marked map; and sending the marked map to a vehicle running in the second road section.

According to an embodiment of the present application, the steps involved in the lane line generation method shown in fig. 3 may be performed by respective modules in the lane line generation apparatus shown in fig. 10. For example, step S101 shown in fig. 3 may be performed by the obtaining module 111 in fig. 10, and step S102 shown in fig. 3 may be performed by the determining module 112 in fig. 10; step S103 shown in fig. 3 may be performed by the extraction module 113 in fig. 10; step S104 shown in fig. 3 may be performed by determination module 112 in fig. 10; step S105 shown in fig. 3 may be performed by the generation module 114 in fig. 10.

Likewise, according to an embodiment of the present application, the steps involved in the lane line generation method shown in fig. 5 may be performed by the respective modules in the lane line generation apparatus shown in fig. 10. For example, step S201 shown in fig. 5 may be performed by the obtaining module 111 in fig. 10, and step S202 shown in fig. 5 may be performed by the determining module 112 in fig. 10; step S203 shown in fig. 5 may be performed by the extraction module 113 in fig. 10; step S204 shown in fig. 5 may be performed by the determination module 112 in fig. 10; steps S205-S207 shown in FIG. 5 can be performed by the generation module 114 in FIG. 10.

According to an embodiment of the present application, each module in the lane line generation apparatus shown in fig. 10 may be respectively or entirely combined into one or several units to form the lane line generation apparatus, or some unit(s) may be further split into at least two sub-units with smaller functions, which may implement the same operation without affecting implementation of technical effects of the embodiment of the present application. The modules are divided based on logic functions, and in practical applications, the functions of one module may also be implemented by at least two units, or the functions of at least two modules may also be implemented by one unit. In other embodiments of the present application, the lane line generating apparatus may also include other units, and in practical applications, these functions may also be implemented by assistance of other units, and may be implemented by cooperation of at least two units.

According to an embodiment of the present application, the lane line generation apparatus shown in fig. 10 may be constructed by running a computer program (including program codes) capable of executing the steps involved in the corresponding method shown in fig. 3 on a general-purpose computer device such as a computer including a processing element such as a Central Processing Unit (CPU), a random access storage medium (RAM), a read-only storage medium (ROM), and a storage element, and implementing the lane line generation method of the embodiment of the present application. The computer program may be recorded on a computer-readable recording medium, for example, and loaded into and executed by the computing apparatus via the computer-readable recording medium.

Fig. 11 is a schematic structural diagram of a computer device according to an embodiment of the present application. As shown in fig. 11, the computer apparatus 1000 may include: the processor 1001, the network interface 1004, and the memory 1005, and the computer apparatus 1000 may further include: a media content interface 1003, and at least one communication bus 1002. Wherein a communication bus 1002 is used to enable connective communication between these components. The media content interface 1003 may include a Display screen (Display) and a Keyboard (Keyboard), and the selectable media content interface 1003 may also include a standard wired interface and a standard wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., W)_I-F_IAn interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., at least one disk memory). The memory 1005 may optionally be at least one memory device located remotely from the processor 1001. As shown in fig. 11, the memory 1005, which is a kind of computer-readable storage medium, may include therein an operating system, a network communication module, a media content interface module, and a device control application program.

In the computer device 1000 shown in fig. 11, the network interface 1004 may provide a network communication function; while the media content interface 1003 is primarily an interface for providing input for media content; and the processor 1001 may be used to invoke a device control application stored in the memory 1005 to implement:

the processor 1001 may be configured to invoke the device control application stored in the memory 1005 to enable determining a side position point of the outline of the virtual side lane line reflecting the second road segment based on the center position point and the width information of the second road segment, including:

Optionally, the processor 1001 may be configured to call the device control application program stored in the memory 1005, so as to determine the side position point for reflecting the outline of the virtual side lane line of the second road segment according to the first average vector and the width information of the second road segment, including:

translating the first average vector to the second center location point;

Optionally, the number of the side position points is at least two; the processor 1001 may be configured to invoke a device control application stored in the memory 1005 to implement generating a virtual side lane line for the second road segment from the side position points, including:

acquiring position information of at least two side position points;

Optionally, the processor 1001 may be configured to invoke the device control application program stored in the memory 1005, so as to perform smoothing processing on the candidate side lane lines, so as to obtain the virtual side lane line of the second road segment, where the method includes:

Optionally, the smoothing processing parameter includes a limit angle between lane line segments; the processor 1001 may be configured to invoke the device control application stored in the memory 1005, so as to perform the smoothing processing on the candidate side lane line according to the smoothing processing parameter, so as to obtain the virtual side lane line of the second road segment, including:

Optionally, the smoothing parameter further includes an iteration limit number of smoothing; the processor 1001 may be configured to invoke the device control application program stored in the memory 1005, so as to implement angle adjustment on the candidate side lane line according to the smoothing parameter if the target angle is greater than the limit angle, so as to obtain the virtual side lane line of the second road segment, including:

the second side position point P_i-1Leveling toward a direction approaching the third side lane segmentMoving to obtain a fourth side position point Q_i-1；

Optionally, the processor 1001 may be configured to invoke a device control application stored in the memory 1005 to implement:

the second side position point P_i-1Translating the lane line segment to the direction close to the third side to obtain a fourth side position point Q_i-1The method comprises the following steps:

determining a center position point on the sixth lateral lane segment;

Optionally, the processor 1001 may be configured to invoke a device control application program stored in the memory 1005, so as to adjust the candidate parameter obtaining model according to the labeled smoothing processing parameter and the predicted smoothing processing parameter, so as to obtain the target parameter obtaining model, where the method includes:

Alternatively, the processor 1001 may be configured to call the device control application stored in the memory 1005 to extract a center position point of the contour reflecting the center lane line of the second road segment from the center lane line of the second road segment, including:

acquiring road section attribute information of the second road section;

Optionally, the first road segment includes a first straight road segment and a second straight road segment; optionally, the processor 1001 may be configured to call the device control application stored in the memory 1005, so as to determine the width information of the second road segment according to the actual side lane line of the first road segment, including:

acquiring a map including the second road section;

marking the virtual side lane line of the second road section in the map to obtain a marked map;

and sending the marked map to a vehicle running in the second road section.

Meanwhile, the candidate side lane lines are subjected to smoothing processing, so that the smoothness of the virtual lane lines of the second road section is improved, the virtual side lane lines of the second road section are closer to the actual side lane lines, and the accuracy of obtaining the side lane lines is improved.

It should be understood that the computer device 1000 described in this embodiment of the present application may perform the description of the lane line generation method in the embodiments corresponding to fig. 3 and fig. 5, and may also perform the description of the lane line generation apparatus in the embodiments corresponding to fig. 10, which is not described herein again. In addition, the beneficial effects of the same method are not described in detail.

Further, here, it is to be noted that: an embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program executed by the lane line generation apparatus mentioned above, and the computer program includes program instructions, and when the processor executes the program instructions, the description of the lane line generation method in the embodiment corresponding to fig. 3 and fig. 5 can be executed, so that details are not repeated here. In addition, the beneficial effects of the same method are not described in detail. For technical details not disclosed in embodiments of the computer-readable storage medium referred to in the present application, reference is made to the description of embodiments of the method of the present application.

As an example, the program instructions described above may be executed on one computer device, or on at least two computer devices located at one site, or on at least two computer devices distributed over at least two sites and interconnected by a communication network, and the at least two computer devices distributed over at least two sites and interconnected by the communication network may constitute a blockchain network.

The computer-readable storage medium may be the lane line generation apparatus provided in any of the foregoing embodiments or an internal storage unit of the computer device, such as a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Memory Card (SMC), a Secure Digital (SD) card, a flash card (flash card), and the like, provided on the computer device. Further, the computer-readable storage medium may also include both an internal storage unit and an external storage device of the computer device. The computer-readable storage medium is used for storing the computer program and other programs and data required by the computer device. The computer readable storage medium may also be used to temporarily store data that has been output or is to be output.

The terms "first," "second," and the like in the description and claims of embodiments of the present application and in the drawings are used for distinguishing between different media content and not for describing a particular order. Furthermore, the terms "comprises" and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or apparatus that comprises a list of steps or elements is not limited to the listed steps or modules, but may alternatively include other steps or modules not listed or inherent to such process, method, apparatus, product, or apparatus.

An embodiment of the present application further provides a computer program product, which includes a computer program/instruction, and when the computer program/instruction is executed by a processor, the description of the lane line generation method in the embodiment corresponding to fig. 4 and fig. 7 is implemented, and therefore, details will not be repeated here. In addition, the beneficial effects of the same method are not described in detail. For technical details not disclosed in the embodiments of the computer program product referred to in the present application, reference is made to the description of the method embodiments of the present application.

Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the examples described in connection with the embodiments disclosed herein may be embodied in electronic hardware, computer software, or combinations of both, and that the components and steps of the examples have been described in a functional general in the foregoing description for the purpose of illustrating clearly the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

The method and the related apparatus provided by the embodiments of the present application are described with reference to the flowchart and/or the structural diagram of the method provided by the embodiments of the present application, and each flow and/or block of the flowchart and/or the structural diagram of the method, and the combination of the flow and/or block in the flowchart and/or the block diagram can be specifically implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable lane line generating device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable lane line generating device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable lane-line generating device to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block or blocks of the block diagram. These computer program instructions may also be loaded onto a computer or other programmable lane-line generating device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer implemented process such that the instructions which execute on the computer or other programmable device provide steps for implementing the functions specified in the flowchart flow or flows and/or block or blocks.

The above disclosure is only for the purpose of illustrating the preferred embodiments of the present application and is not to be construed as limiting the scope of the present application, so that the present application is not limited thereto, and all equivalent variations and modifications can be made to the present application.

Claims

1. A lane line generation method, comprising:

2. The method of claim 1, wherein the center location points comprise a first center location point, a second center location point, and a third center location point; the first central position point and the second central position point have an adjacent relation, and the second central position point and the third central position point have an adjacent relation;

determining a side position point for reflecting the outline of the virtual side lane line of the second road section according to the center position point and the width information of the second road section, including:

3. The method of claim 2, wherein determining the side position point of the contour reflecting the virtual side lane line of the second road segment based on the first average vector and the width information of the second road segment comprises:

translating the first average vector to the second center location point;

4. The method of any of claims 1-3, wherein the number of side position points is at least two; the generating of the virtual side lane line of the second road section according to the side position point includes:

acquiring position information of at least two side position points;

5. The method of claim 4, wherein smoothing the candidate side lane lines to obtain the virtual side lane line for the second road segment comprises:

6. The method of claim 5, wherein the smoothing parameters include a limit angle between lane line segments; the smoothing of the candidate side lane line according to the smoothing parameter to obtain the virtual side lane line of the second road section includes:

7. The method of claim 6, wherein the smoothing parameters further include a number of iteration limits for smoothing;

if the target angle is greater than the limiting angle, performing angle adjustment on the candidate side lane line according to the smoothing processing parameter to obtain a virtual side lane line of the second road section, including:

8. The method of claim 7, wherein said second side position point P is identified_i-1Translating the lane line segment to the direction close to the third side to obtain a fourth side position point Q_i-1The method comprises the following steps:

determining a center position point on the sixth lateral lane segment;

9. The method of claim 5, wherein the method further comprises:

10. The method of claim 9, wherein said adjusting the candidate parameter acquisition model according to the annotated smoothing parameter and the predicted smoothing parameter to obtain the target parameter acquisition model comprises:

11. The method according to claim 1, wherein the extracting, from the center lane line of the second road segment, a center position point of the contour reflecting the center lane line of the second road segment includes:

acquiring road section attribute information of the second road section;

12. The method of claim 1, wherein the first road segment comprises a first straight road segment and a second straight road segment; determining the width information of the second road section according to the actual side lane line of the first road section comprises the following steps:

13. The method of claim 1, wherein the method further comprises:

acquiring a map including the second road section;

and sending the marked map to a vehicle running in the second road section.

14. A lane line generation device, comprising:

15. A computer device, comprising: a processor and a memory;

the processor is connected with the memory; the memory is for storing program code, and the processor is for calling the program code to perform the method of any of claims 1 to 13.